Abstract
Academics have conflicts over ideas with some regularity, yet they contest only some of them. When will they do so? We draw on a theory of conflict management developed by Donald Black and others to explain the response to intellectual conflict. Drawing on interviews with 70 professors at two universities, we find that the contestation of intellectual conflicts is predicted by their social geometry. Academics are more likely to contest conflicts over the validity, ownership, and production of ideas when the conflict spans greater distances in relational and functional space, originates from a lower elevation in vertical space, and is a larger actual or potential change in vertical space.
Conflict—the clash of right and wrong—pervades social life. A universal human experience, conflict occurs in every corner of the social universe. Conflict touches men and women, young and old, rich and poor, powerful and powerless, persons and groups, corporations and nations. Particular episodes or cases exhibit variable intensity: While many conflicts die on the branch, the occasional controversy burgeons into all-out warfare. Conflicts tend to be greedy, devouring significant amounts of time, energy, and resources, individual and institutional. Many have a substantial afterlife, lingering long in the memory of the parties, especially the aggrieved. Who does not recall an assault by a stranger, a public insult by a colleague, a snub by a high-status actor? Conflicts can be profoundly consequential, personally and socially.
Intellectual life is no exception. The production and dissemination of ideas is permeated by conflict. Occasionally, such controversies become well known. Can string theory unite the theories of gravitation and quantum mechanics, or is it an untestable scientific dead end (e.g., Smolin 2006)? How much credit do scholars other than Charles Darwin deserve for discovering the theory of evolution by natural selection (e.g., Stott 2012)? Should the government fund research on human sexual behavior (e.g., Kaiser 2003)? These disputes arose over ideas rather than personal matters (e.g., harassment), professional matters (e.g., teaching obligations), or material matters (e.g., salary). Apart from occurring in different fields, they diverge in certain respects. The debate over string theory is a conflict over the merits of a set of ideas—a validity conflict. The controversy over the theory of evolution centered around who should receive credit for ideas—a creativity conflict. The federal ban on funding sex research sought by some legislators would have blocked the pipeline of ideas—a productivity conflict. Yet they all concern the production and dissemination of ideas—they are intellectual conflicts.
The three controversies all spilled over into the public sphere. Consequently, much was learned about what the protagonists and their supporters did in response to the conflict. In that respect, they are atypical. Most intellectual conflicts do not see the light of day but remain confined to the parties and their intimates and close associates. What might be called nonpublic or private intellectual conflicts are nonetheless important in intellectual life, triggering countless discussions and sapping precious time and energy from scholars. Yet, surprisingly little is known about how academics handle private intellectual conflicts.
We seek to begin to fill that lacuna with data drawn from interviews with 70 faculty members at two universities, one a large state school, the other a medium-sized private university. At both institutions, the types of intellectual conflicts found in the public realm are recapitulated in the private realm. Academics engaged in private intellectual conflict similarly argue over issues of validity (Which ideas are right?), as when our respondent, Abe, was strongly criticized by Ben at a conference for a talk favorable to Charles, a rival of Ben’s. Creativity conflicts (Who deserves credit for the idea?) are relatively common too: One of our respondents reported still being “mad” at being left off the author list of a paper published 40 years earlier. And productivity issues surface with some regularity as well, as when editors turn down papers for publication or collaborators fail to deliver their promised contributions. Of the many possible questions these conflicts raise, we pose here a central one: When will academics contest and when will they tolerate intellectual conflict?
The answer might be straightforward. Contestation depends on the personality of the aggrieved individual: Some people are more assertive and aggressive than others in pursuing their intellectual interests. However, an alternative answer lies in the social environment. In developing our argument, we draw on the most developed theory of conflict management currently available in the sociological literature: that developed by Donald Black (1976, 1995, 2011) and others. Blackian theory explains the handling of conflicts with their location, direction, and movement in social space—their social structure or geometry. While the theory has been applied to diverse forms of conflict, it has never been adapted to the handling of intellectual conflict. Here we employ Blackian theory to reveal several major patterns underlying the contestation of intellectual conflicts.
Theoretical Perspective
Our focus is the handling of, or response to, intellectual conflict. We do not address the outcome of intellectual conflict, such as how discoveries are credited to single or multiple authors (Cozzens 1989) or, more generally, how conflict advances or impedes the growth of knowledge (e.g., Abbott 2001; Collins 1998). Our approach is similarly distinct from the important body of work by Robert Merton (1973) and those working in his tradition.
Mertonian Theory
Merton’s scholarship has addressed intellectual conflict primarily by focusing on how one species of academics—scientists—sometimes deviates from the norms by which their enterprise is governed, such as universalism, communalism, disinterestedness, originality, and skepticism (see e.g., Braxton and Bayer 1994; Braxton, Proper, and Bayer 2011; Zuckerman 1988). 1 Although deviance is the main focus of Mertonian work, the response to deviance (conflict management) has not been wholly overlooked (see e.g., Braxton 1999). Discussing charges of plagiarism, Merton (1957:653) mentions the “familiar deterioration of standards governing conflictual interaction. . . . Reinforced by group loyalties and often by chauvinism, the controversy gains force, mutual recriminations of plagiary abound, and there develops an atmosphere of thoroughgoing hostility and mutual distrust” (see also Hagstrom 1965). Not addressed, however, is variation in the handling of conflict. Under what conditions does the aggrieved party go to war, grumble to a friend, or ignore the violation entirely? For instance, Merton’s (1968) well-known paper on the Matthew effect documents the recognition conferred by the scientific community on higher status scientists but does not analyze the reaction of their lower status collaborators or colleagues who received a diminished share of the plaudits for the same discovery. Braxton (1991) reports that departmental chairs tend to handle breaches of Mertonian norms of misconduct by their colleagues informally (e.g., making disapproving comments). In a survey of senior chemists, Braxton and Bayer (1996) found that those who had come across research misconduct were likely to respond informally when they believed that taking action could adversely affect their career or publicly embarrass the offender. But the study does not examine other possible influences. For a body of ideas that may explain the handling of intellectual conflict more extensively, we turn to Blackian theory.
Blackian Theory
Blackian work on conflict is rooted in a larger theoretical system that can in principle be extended to any form of behavior, including art, religion, medicine, and ideas (Black 1979, 1995, 2000). The system explains behavior in a novel manner: with its location, direction, and movement in social space or its social geometry (Black 1976, 1995). Social space has multiple dimensions, including the horizontal (the distribution of relationships), organizational (the capacity for collective action), cultural (the distribution of symbols), normative (the definition and response to deviance), and vertical (the distribution of wealth). Each dimension is a matter of degree, and together they incorporate the principal variables of prior theoretical traditions (e.g., social class, the division of labor, bureaucracy). 2
Social space has a synchronic (stationary) and a diachronic (dynamic) aspect. The synchronic aspect consists of the relative location of the parties in social space—for example, their relational closeness—and the direction of the interaction in social space—whether the encounter is directed upwardly, downwardly, or laterally. The diachronic aspect is a change of the parties’ location in social space—such as an increase or decrease in intimacy or a rise or fall in status. Both help to explain the response to conflict.
The synchronic aspect of social space was initially developed by Black (1976) in his work on the handling of legal conflict. Subsequently, Black and others extended geometric theory beyond law to explain a wide variety of conflict management patterns (Black 1993; see also Campbell 2011; Cooney 2009b). Forms of conflict management addressed include avoidance, criticism, apology, and therapy (Baumgartner 1988, 1998; Cooney and Phillips 2012; Hoffmann 2006; Horwitz 1982). Also theorized has been the handling of conflict in a variety of settings, including drug markets, public places, flat corporations, executive offices, and suburbia (Baumgartner 1988; Cooney 2009a; Jacques and Wright 2008; Morrill 1991, 1995; Tucker 1999). Particular attention has been paid to the social geometry of various forms of violence, from suicide and interpersonal homicide to collective violence (rioting, lynching, terrorism, vigilantism, and genocide) (Black 2004; Campbell 2009, 2015; Cooney 1998; Manning 2012; Phillips 2003; Phillips and Cooney 2005; Senechal de la Roche 1996). Thus, violent responses to conflict become more likely and severe when parties are relationally distant (see e.g., Phillips 2003). For the same conflict, people are more likely to launch physical attacks on strangers than intimates. Intimate violence typically occurs after a longer history of provocations and disputes. Functional distance similarly raises the probability of violence (see e.g., Senechal de la Roche 1996; see also Phillips 2003). Tension between individuals, groups, and nations is more likely to take a violent turn when the parties do not depend on one another for survival. Relative status is also important. Actors tend to be more forgiving of misconduct by those above them in the social hierarchy than those below them: Social superiority attracts assertiveness while inferiority attracts restraint (see e.g., Campbell 2009). While intellectual conflicts rarely become violent within the academy, they too differ in how assertively parties respond to them. Does the handling of private intellectual conflict vary with its social geometry in similar fashion?
Black’s 2011 book Moral Time expanded the concept of social space to include a diachronic aspect. Black proposes that conflict is a direct function of change in social space (or what he terms movements of social time). Three categories of change are of particular importance: increases or decreases in inequality (vertical time), increases or decreases in intimacy (relational time), and increases or decreases in diversity (cultural time). The same act may involve more than one movement of social time. A physical assault, for example, is a significant invasion of personal space, an increase in physical intimacy, and therefore a large movement of relational time. Assault, in addition, may injure its victim, damaging the fundamental means of production and form of wealth: the human body. As such, it is a loss of status, a movement of vertical time. The greater and faster the movement of social time, the more conflict it causes.
Since the diachronic conception of social space is relatively new, it has generated comparatively little scholarship. Even so, the idea of social time has been invoked to explain the causes and handling of several aspects of conflict: suicide, online hostility, family honor killing, and genocide (Campbell 2013; Cooney 2014; Manning 2014; Phillips and Cooney 2015).
All three spatial changes can trigger and intensify intellectual conflict. Thus, the forming or breaking of relationships may result in contested conflict over ideas, as may the meeting or rejection of people of different cultures. However, the dominant source of contested conflict over ideas is likely to be change in inequality: movements, actual or threatened, up and down the status ladder. Hierarchy is central to the social organization of the American academy with its formal and informal ranking of actors (students, assistant professors, associate professors, full professors, superstar academics). Academics compete to publish in the best journals and with the most prestigious presses, acquire more and larger grants, and accumulate the highest honors and prizes. In this race to succeed, some win, and others lose. The hierarchy is unstable as tenure may be denied, promotion postponed, or prize seeking thwarted. A surge in publications may summersault a person above his or her peers while a lapse in productivity may send his or her reputation plummeting. Conflicts that alter status in a field, actually or potentially, such as having an idea stolen or an important publication blocked, appear to be particularly likely to result in conflict.
The exact relationship between the synchronic and diachronic aspects of social space is still being explored. Here we combine both aspects to propose that the handling of intellectual conflict is a product of all three components of social geometry—location, direction, and change in social space. We measure location with relational distance and functional distance, direction with status superiority, and change with the movement of vertical time. We begin with location in social space. But first, a word about our study.
The Study
Our study has two components: formal and informal. The formal part consists of interviews with a random sample of 70 professors at two American universities drawn from all major academic fields. After conducting 15 pilot interviews with friends to refine the instrument, we conducted the formal research interviews at a medium-sized private university in the West and a large state university in the Southeast. Both are ranked in the broad middle of national universities (i.e., the 50–100 range by US News and World Report). We acquired a list of all the faculty members at each university and narrowed the list to eligible respondents: full-time tenure-track and tenured professors who conduct research (thus, we eliminated artists, performers, and administrators above the level of department chair who do not focus on research; we also eliminated members of our own departments given the sensitive nature of the topic). We then drew a random sample of respondents at each university. We sent an invitation letter through campus mail and followed up with an email to ask whether the person was interested in participating. We repeated the procedure—drawing a random sample and recruiting participants—until we reached our goal of 35 usable interviews at each school.
We conducted the interviews over the summer, reasoning that faculty would be less inclined to meet during the academic year when they have more time-sensitive obligations. The disadvantage of the summer, however, is that many faculty are traveling or do not want to have their precious research time interrupted. Of the 140 faculty members randomly selected and recruited at the private university, 45 agreed to be interviewed, 38 completed the interview, and 35 had experienced intellectual conflict; of the 134 faculty members randomly selected and recruited at the state university, 44 agreed to be interviewed, 39 completed the interview, and 35 had experienced intellectual conflict. The 70 respondents represent all academic ranks (24 percent assistant professors, 24 percent associate professors, and 51 percent professors) and a broad range of areas (41 percent social sciences, 29 percent physical sciences, 19 percent humanities, and 11 percent math/statistics/engineering/computer science). The majority were men (74 percent). 3
Our invitation letter stated that respondents would be asked to describe the most serious intellectual conflicts they had experienced during their careers (from graduate school to the present). Most respondents came prepared with a mental or written list. Perhaps surprisingly, respondents often provided animated and detailed descriptions of incidents that occurred decades ago: Serious intellectual conflicts may leave an indelible mark. As the interview came to a close, we also described common forms of intellectual conflict that other respondents had mentioned in order to “shake the bushes,” occasionally prompting the respondent to recall another incident. In total, the 70 respondents described 270 intellectual conflicts (ranging from 1 to 13 with a mean of 4). Because the unit of analysis is the conflict—not the respondent—the final sample size is 270 conflicts. 4
The interviews were conducted in the respondent’s campus office and lasted about 2 hours (a few were conducted in the researcher’s office or the home of the researcher/respondent). We used the recorded interviews to summarize each conflict (by transforming circuitous stories into sequential narratives), establish the social geometry of each conflict, and examine whether the respondent contested the grievances. 5
Two universities cannot represent the thousands of colleges and universities in the United States. However, our unit of analysis is not universities or even faculty members but rather intellectual conflicts. Sampling intellectual conflicts from informants at a medium-sized private university and a large state university in different parts of the country offers a reasonable starting point for the initial study on the topic. Because the vast majority of intellectual work is done in such settings, we presume that the vast majority of intellectual conflicts occur in such settings.
Our formal interviews are supplemented by informal observations and discussions with colleagues conducted over many years. Many of the discussions were casual conversations in the department, at conferences, or when visiting other universities. Unsystematic as it was, the informal ethnography yields a set of real-world conflicts from a 15-year period. Those incidents complement the cases produced more methodically by the formal interviews. Together, they provide a body of illustrative but by no means definitive information on a topic much discussed but rarely studied: the handling of conflict between academic intellectuals.
We have concealed the identity of the participants. We use names beginning with the letter A to refer to the respondent, and those beginning with B and C refer to the person(s) the respondent had a grievance against. In some cases, we have also changed the respondent’s discipline (e.g., from an anthropologist to a chemist) and/or the origin of the conflict (e.g., from fabricating experimental data to fabricating survey data). Occasionally, we changed the respondent’s gender. We also limited the type of detail and nuance that one might expect in a qualitative description of a case because adding flesh and bones might allow the reader to identify the parties; our respondents and readers are both academics who could know one another.
The Geometry of Contestation
Common wisdom within academia states that certain individuals are simply more intellectually aggressive than others. And there is surely some truth in that. Some people seem to enjoy the cut and thrust of intellectual combat; others shy away from it. Personality may therefore explain contestation. Our respondents certainly thought so. When asked why they believed some conflicts were pursued and others tolerated, the vast majority considered the personality of the aggrieved to be the key factor. In fact, personality appears to be a relatively weak predictor as none of our respondents was invariably confrontational. More telling was the situation the respondent encountered or, more precisely, the configuration of the parties in social space—the relational distance, functional distance, and vertical direction of the conflict coupled with the movement of vertical time.
Relational Distance
Relational distance is a degree of participation in the life of another (Black 1976). Intellectual conflicts do not escalate easily over short distances in relational space. Toleration of intimates may extend even to egregious wrongs, as in the case of Alfred and Bud, close friends from graduate school. After they graduated, Bud came to stay with Alfred and inquired what he was working on. Alfred replied that he wanted to apply a new philosophical system to the field and went on to describe the project he had in mind at some length. The following year, Alfred was shocked when Bud published a book applying the same philosophical system to the same field without any attributions to Alfred. In addition, Alfred suspects that Bud tried to undermine his work when his own book on the subject appeared. Despite harboring these strong grievances, Alfred never confronted Bud, still talks to him a few times a year, and even continues to regard him as a friend.
As relational distance increases, intellectual conflicts become more confrontational. Indeed, the same person may be more or less likely to stand up to the same adversary as their relational distance waxes or wanes. Hostile in print, polite in person is not a rare combination, for instance. After Barry critiqued Alex’s work in a prominent journal, the editor allowed Alex to publish a short rebuttal that he worded sharply. When the two attended a European conference a year or so later, they happened to stay in the same hotel. They greeted each other warmly, dined together several times, and visited some city sights together. Neither man ever mentioned the sharp exchange. Similarly, verbal criticism is often more muted in person. A big-name political scientist, Brandon, once tried to recruit Angus, another big-name political scientist, to join an organization that he was forming to promote the theoretical perspective for which Brandon was well known. Declining politely, Alex told Brandon that he was “a poor candidate for conversion.” In private, Angus is considerably more scathing about Brandon’s perspective.
In a triad, A and B may be closer than A and C. Conflict between A and B may therefore be less confrontational than conflict between A and C over the same issue. In one case, Bree invited a departmental colleague with strong methodological skills, Agnes, to join in a collaboration she had initiated with Carmel, a scholar in another university. Reading the paper, Agnes concluded that it provided a mere literature review. Agnes came up with a new frame, research questions, and data for simulation models. She ran the simulations and wrote the paper, using some of the original literature review. Having done the lion’s share of the work, Agnes believed she should be first author on the paper. Her departmental colleague, Bree, agreed. But the other collaborator, Carmel, whom she barely knew, demurred, arguing that the original idea for the paper was hers and Bree’s. Eventually, they compromised by listing the authors in alphabetical order, a resolution that continues to irritate Agnes.
More relational distance attracts more contestation. When an article appeared in an international magazine on a topic he had pioneered, Apollo acted forcefully to defend his authorship. The article did not mention Apollo’s work directly but only cited a newspaper writer who had published an article summarizing his work. Stung by the elision of his contribution, Apollo wrote a strongly worded letter to the editor of the magazine, whom he did not know. Yet Apollo regularly co-publishes with a friend whom, he feels, does not pull his weight on their projects. Although resentful of the free riding, Apollo has never approached his friend about removing his name from any of their papers.
Functional Distance
Just as relational distance increases the willingness to contest, so too does functional distance—the degree to which the parties are independent of each other for material, intellectual, or other resources. The greater their independence, the more confrontational intellectuals are; conversely, the more dependent they are on the other party, the more tolerant they tend to be.
Functional distance varies by field. 6 In large fields, the availability of alternate outlets may encourage authors to be assertive in, for example, questioning the decisions of reviewers and editors. By contrast, small academic specialties often have a few core actors who exert considerable influence over what gets published in the best journals. Newcomers may have to proceed with particular caution in order not to alienate the gatekeepers. Consequently, when an individual migrates from a large to a small field, his or her tendency to contest adverse decisions may change as well. For example, Astrid is a biologist who initially worked within a large branch of the discipline. On several occasions, she contested the rejection of a paper and once even managed to persuade an editor to reverse a negative decision. In recent years, Astrid has entered a smaller, more applied subfield. Even though she is now more senior in her career, she has become more reluctant to question reviewers’ and editors’ decisions. In one case, the editor of the top journal sat on a paper of hers for two years. Astrid tolerated the frustrating delay, sensing that if she made a fuss she might never get published in the future.
Of all the dependent relationships in academia, a graduate student’s reliance on a major professor is perhaps the most extreme. In one case, Aeneas had to tolerate seeing Cindy’s name go before his on the author list of several papers he had largely written and to which Cindy had contributed virtually nothing. But Cindy was going up for promotion to full professor, and Bruce, her senior full professor spouse and the holder of the grant from which the publication data were generated, insisted she be placed ahead of Aeneas. The sheer injustice galled Aeneas. But he was stuck: He needed Bruce’s data, and it was too late to switch to another major professor. He therefore put up with what he considered to be theft of his work.
Another familiar form of dependence in American universities arises when junior faculty members need the votes of more senior members of their department to receive tenure and promotion. Dependent juniors will often tolerate conduct from their more senior colleagues that they consider wrong. When Aileen was an assistant professor, she was approached by a more senior departmental colleague to collaborate. Aileen wrote the paper and got a revise and resubmit from a journal. Her colleague volunteered to handle the revisions. When the paper was subsequently rejected, Aileen discovered that the reviewers had queried how the sample size of 300 could increase to 500 in some of the analyses. She suspected that her colleagues had made up the data. When Aileen gently raised the issue of the discrepant numbers, he dodged the question and said he would get back to her. He never did. Although she had put considerable work into writing the original version, Aileen did not want to rock the boat: She knew her colleague would soon be voting on her tenure application, so she let the project—and his possible fabrication of data—die quietly.
After tenure, academics may remain dependent on particular individuals in other capacities. For instance, as associate editor of a journal, Adriana declined to send a paper out for review on the grounds that it did not fit the subject matter of the journal. What caused Adriana to think hard about her decision was that the paper was written by an acquaintance, Connie, herself a former associate editor of the journal. When Connie got nowhere complaining to Adriana, she appealed to the editor, Brett, who happened to be a close friend. Brett reversed Adriana’s decision and accepted Connie’s paper, a decision that Adriana still feels strongly was wrong and for the wrong reasons. But she simply tolerated the decision. She was fairly new in the job and “wanted to be in Brett’s good graces because he is a senior person in our field and he’s the kind of person you want as your friend, you don’t want as an enemy.”
Vertical Superiority
Vertical superiority is the extent to which the aggrieved party is higher in status than the adversary (e.g., of higher rank, more senior, more cited). Holding constant the conduct, the greater the social superiority, the more likely the aggrieved party is to contest. Conversely, the more inferior the aggrieved party, the less likely confrontation becomes. When Ainsley was in graduate school, she was one of several teaching assistants in a large class. The professor would meet with the graduate students each week, listen to their ideas, and then use those ideas in subsequent classes as if they were his own. Since the graduate students attended the class, the expropriation took place in front of the very people he expropriated. So blatant was the theft that the graduate students joked about it among themselves. They believed the professor must be “pathological or stupid” or else “felt he has so much power that he doesn’t care.” Despite the repeated victimization and their unanimous condemnation of it, none of the students challenged the professor either individually or collectively.
Vertical superiority and functional distance may march together in that status inferiors are often dependent as well. But superiority exerts an influence separately of independence. Alistair, an assistant professor, had a grant with a senior professor, Brent, who had technical resources that were helpful for the grant. Shortly afterward, Brent left the department and took an administrative position without informing Alistair, dumping all the grant work on him. The result was that Alistair had to extend the work out two additional years. Although annoyed and inconvenienced, Alistair never voiced his complaint to Brent, even though he had little ability to adversely affect Alistair’s career as he was no longer in the field.
When inferiors do confront superiors, they tend to plan it out carefully. Andrea devoted considerable thought to figuring out how she would handle her senior co-author, Barbara, who did not use the theoretical framework they had agreed on for their paper. The omission upset Andrea greatly, who vented to her husband about the problem. When Andrea eventually tentatively confronted Barbara, she was relieved to discover her open to guidance.
Upward confrontation is not just deliberate but often muted as well. After noticing that a senior professor at an Ivy League school, Brooke, was working on the same problem as Carol, one of his graduate students, Andy arranged a meeting with Brooke, and they agreed to collaborate on a paper. Brooke sent her data to Andy. Carol did most of the work on the paper, and she and Andy sent it to Brooke with all three of them listed as co-authors. Brooke objected to the analysis, however, and would not allow the paper to be submitted for publication, though she would not do anything to improve the paper. Annoyed, Andy complained to Brooke but admits he did not do so very strongly. Compared to him, Brooke, he said, was “a big shot.” If she did not think the paper good enough, who was he to insist it was or demand she improve it?
Grievances in a downward direction, however, are more forcefully pursued. A senior figure in her field, Amy had a paper rejected by a top journal without even being sent out for review. The associate editor handling the paper said that it was not the kind of work they publish. The decision infuriated Amy. She knew the paper was appropriate and indeed probably good enough to be published in the journal. She complained to the associate editor, who went to the editor. He gave it to another associate editor, who sent it out for review. Interestingly, earlier in her career Amy had faced a similar problem but in an upward direction and had handled it much less assertively. She received a highly negative review of a book manuscript she had submitted to a prestigious university press. The review, she believes, was almost certainly written by a senior figure known for protecting his intellectual territory. Although she considered the review unfair and inaccurate, she did not protest it but simply took her manuscript to a lower status press.
Vertical Time
Vertical time, recall, is any shift in inequality, actual or threatened. 7 The larger and faster the movement of vertical time, the more likely an incident is to be contested, regardless of whether it occurs among intimates or strangers, independents or dependents, superiors or inferiors.
In academic life, the size of the movement of vertical time can be measured by the extent of the threat to a career. To be embarrassed, as when our respondent, Arnold, was sharply criticized on a conference panel, is a relatively small movement. To experience a setback that does not seriously damage a career, as when Ava received incompetent reviews on a paper, is a medium movement. To fail a pre-tenure review or have a central publication in a promotion file held up while a reviewer demands ever-greater changes—as happened to two of our respondents—has the potential to greatly damage a career and is therefore a large movement of vertical time.
A public accusation of plagiarism can also have very deleterious career consequences for academics. Equally, to be the victim of plagiarism or idea theft more generally can be devastating, depending, for example, on the importance of the idea and the unequivocality of the intellectual larceny. Abel and Bart were close friends in different social science disciplines who shared an academic interest in the family. Bart published in other areas but taught a course on the family; Abel had written a book on the topic. Bart shared his syllabus with Abel. Later, Abel sent a draft of a paper to Bart and was shocked to receive an email in which Bart claimed that one section of the paper mirrored very closely the organization of his syllabus. In response, Abel sent an initial short email stating that the paper was his own work only, following it up with a longer email in which he defended himself strongly and pointed to six books that used a similar structure as Bart’s syllabus. Although Bart eventually acknowledged that his syllabus was not unique, Abel has remained wary of sharing ideas with Bart for fear of future allegations of plagiarism. In a second case, Abel was on the other side of the conflict, sharing a draft paper with Bryson, a visiting speaker. Some months later, he received a draft paper from Bryson on the same subject. While Bryson’s paper contained an acknowledgment of Abel’s help, Abel considered the paper to be very close to his own. However, the possible loss of status was relatively low as Abel discovered that his own paper would be published first and that Bryson’s paper would be published in a separate discipline with an acknowledgment that would allow readers to trace the idea to him. Hence, he did not contest the conflict.
Since publications are the coin of the academic realm, to be omitted from the author list is a movement of vertical time. The size of the movement depends on the number and visibility of the publications and the status of the author: Each publication means more to an assistant professor’s career than a full professor’s career. Thus, Amos, an assistant professor, was dismayed to learn that a senior colleague, Bradley, with whom he had collaborated at his prior university, was about to publish work from their collaboration without including him as an author. Although Amos was very uncomfortable doing so, he called up Bradley, who told him that his contribution was not important enough to warrant authorship. Digging his heels in, Amos pointed out that without his contribution, Bradley would have nothing to publish. Bradley countered that the work had progressed way beyond where it was when Amos left the laboratory. Bradley eventually agreed to consult with his postdoc, who sided with Amos. In the end, Amos got four crucial publications from the collaboration.
Vertical time, then, along with relational distance, functional distance, and vertical superiority appear to be central to the handling of intellectual conflict. But do they each matter individually when considered collectively? And do they matter when other possible influences on contestation are analyzed? To answer these questions and extend our analysis, we conducted a multivariate analysis of our interview data.
Quantitative Analysis
Although the effects of social geometry appear to be strong, it is nonetheless possible that they are overwhelmed by other aspects of the conflict. Our interviews contained questions about disputatiousness and several additional variables that might exert a far more profound influence on which conflicts are contested. A quantitative analysis allows us to address to what extent they explain away the effects of social geometry.
Measurement
Table 1 describes measurement strategies and descriptive statistics for our three synchronic variables (relational distance, functional distance, and vertical superiority), our diachronic variable (vertical time), and the control variables (each is measured at the time of the conflict). 8 The letters A and B represent the respondent and the person the respondent had a grievance against, respectively.
Measurement Strategies and Descriptive Statistics for the Theoretical Variables and Controls (n = 270 Intellectual Conflicts).
Note: The letters A and B represent the respondent and the person the respondent had a grievance against, respectively.
To measure relational distance, we asked an open-ended question about A’s relationship with B. Based on the responses, we created an ordinal scale focused on friendship. From the lowest level to the highest level of relational distance, the categories are: friends who work in the same department, friends who work in different departments, and those who are not friends.
Our measure of functional distance is based on an open-ended question regarding the extent to which B was in a position to influence A’s success in academia. We developed a rubric containing two elements: how much B could influence A’s career (high impact such as a dissertation advisor or tenure reviewer, medium impact such as a grant reviewer, low impact such as a journal reviewer, and no possible impact) and how often B had done so (several occasions and future potential or a few occasions already) (see the end of Table 1). Also an ordinal variable, functional distance was coded according to the cell of the rubric that reflected the degree of reliance, with the categories being: extremely dependent, significantly dependent, partly dependent, and independent. 9 Note, too, that while vertical superiority may overlap with functional distance, the theoretical concepts are distinct: Academics are not necessarily dependent on those above them in rank, and those who are dependent are not necessarily inferior in rank to those on whom they depend.
Vertical superiority is based on the ranks reported by our respondents: graduate student, liminal position (e.g., postdoc, lecturer, medical resident), assistant professor, associate professor, professor, and professor who is a leader in the field (university administrators are included in the highest rank because most are professors with heightened organizational status). We created a three-category ordinal measure: superior (A is one or more ranks above B), equal (A and B are the same rank), and inferior (A is one or more ranks below B).
Finally, to measure the movement of vertical time, we created an ordinal scale that captures the potential/actual consequences of the conflict for the respondent’s career. A small movement of vertical time occurred when the respondent was frustrated or embarrassed but his or her career was not truly diminished (e.g., being harshly criticized on a conference panel). A medium movement of vertical time occurred when the respondent’s career was diminished but not truly damaged (e.g., having a paper rejected because of incompetent reviews). A large movement of vertical time occurred when the respondent’s career was truly damaged (e.g., being denied tenure).
Our dependent variable is contestation. To contest is to dispute—to press, rebut, and perhaps even quarrel (as opposed to ignoring the grievance or raising the grievance in a subtle, gentle, or indirect manner yet declining to push). 10 Contestation can be offensive (A pursued the grievance against B) or defensive (A responded to B’s pursuit of a grievance). Of the 270 intellectual conflicts, 112 were contested. 11
To control for potential confounding variables, we examined four commonsense explanations for contestation: anger, disputatiousness, type of intellectual conflict, and tenure. We asked respondents to rate their anger about the conflict on a scale from 1 to 10. We also asked respondents to rate their disputatiousness (general approach to confrontation at work) on a scale from 1 to 10, an attempt to capture assertiveness. We included the type of intellectual conflict (validity, creativity, productivity) under the assumption that respondents would be more concerned about receiving credit for ideas and defending the truth of ideas than the more routine production of ideas. Finally, we considered the respondent’s tenure status as the protection afforded by tenure might promote assertiveness.
We transformed the original 1 to 10 scales for anger and disputatiousness into dummies because a small number of respondents preferred words over numbers. A respondent who was reluctant to choose a number on the anger scale might say: “I wasn’t mad,” or “I was sort of upset,” or “I was furious.” A respondent who was reluctant to choose a number on the disputatiousness scale might say: “I almost never confront people at work,” or “I choose my battles,” or “If I’m mad the person is definitely going to know about it because I confront situations head on.” To retain these cases, we focused on the respondents who did provide numbers, coding the top quartile of scores as high anger and high disputatiousness, the middle 50 percent of scores as medium anger and medium disputatiousness, and the bottom quartile of scores as low anger and low disputatiousness. We then matched words to the categories created with numbers.
In addition to controlling for commonsense explanations, we also controlled for the gender configuration of the conflict (female-male, female-female, male-female, or male-male) 12 and the race of the respondent (non-white or white). Because only 20 of the 270 conflicts came from non-white respondents, the findings for race must be considered tentative.
We used logistic regression to examine whether the social geometry of an intellectual conflict predicted contestation (1 = contest; 0 = not). Robust standard errors were estimated in Stata to adjust for the clustering of intellectual conflicts within respondents (fixed effects models were also estimated on a subset of the data as a sensitivity test; see note 14).
Results
We begin by examining bivariate relationships. The cross-tabulations in Figure 1 demonstrate strong and significant linear relationships for each of our four geometric variables (each relationship is significant at p ≤ .05).
Bivariate patterns: contestation by social geometry (n = 270).
To determine whether such patterns hold after controlling for confounding variables, Table 2 presents odds ratios from the logistic regression of contestation on social geometry. Models 1 through 4 consider the theoretical variables separately. Models 1 and 2 demonstrate that respondents were more willing to contest intellectual conflicts across greater expanses of social distance: Each unit increase in relational distance raised the odds of contestation by 79 percent, and each unit increase in functional distance raised the odds of contestation by 56 percent. As Model 3 reveals, each unit increase in vertical superiority—from inferior to equal to superior—increased the odds of contestation by 64 percent. Model 4 indicates that each unit increase in the movement of vertical time increased the odds of contestation by 47 percent. Model 5 considers all the theoretical variables simultaneously (with controls). Here, one might expect the initial relationships to disappear as the synchronic elements of social geometry overlap. Superiority and independence often go hand-in-hand (78 percent of superiors were independent, compared to 48 percent of equals and 18 percent of inferiors). Similarly, intimacy tends to flourish among equals (29 percent of equals were friends, compared to 16 percent of superiors and 14 percent of inferiors). Yet the odds ratios remain robust and significant in the full model: for each unit increase in relational distance, functional distance, vertical superiority, and vertical time the odds of contestation increase by a factor of 1.8, 1.5, 1.7, and 1.8, respectively. 13 In sum, the multivariate models suggest that the theoretical variables predict contestation separately and in combination.
Multivariate Patterns: Odds Ratios from the Logistic Regression of Contestation on Social Geometry (Robust Standard Errors Used to Adjust for the Clustering of Intellectual Conflicts within Individuals) (n = 270 Intellectual Conflicts).
Note: The letters A and B represent the respondent and the person the respondent had a grievance against, respectively.
p ≤ .10. **p ≤ .05. ***p ≤ .01.
To explore further how critical social geometry is to the handling of intellectual conflict, we calculated predicted probabilities based on the coefficients in Model 5 with controls held constant at the mean. Consider the following hypothetical yet familiar scenarios. In Scenario I, a graduate student has a conflict with her major professor. The graduate student who experienced a small movement of vertical time is inferior, extremely dependent, and has developed a friendship with her advisor who works down the hall. Consequently, the predicted probability of contestation is a mere .05. In Scenario II, an assistant professor has a conflict with another assistant professor who works at a different university. After meeting on a conference panel, the two became friends. Because both work on the same topic, they often serve as reviewers for each other on papers and grants. Here, the assistant professor who experienced a moderate movement of vertical time is a status equal in a position of significant dependence, so the predicted probability of contestation remains relatively low but nonetheless rises to .29. Finally, in Scenario III, a full professor has a conflict with an assistant professor at a different university who accused her of plagiarism. The full professor who experienced a large movement of vertical time is independent of and superior to the stranger who is the object of her ire. As a result, the predicted probability of contestation catapults to .83. While these predicted probabilities imply a higher level of precision than the data truly warrant, they make plain nonetheless our central finding: The contestation of intellectual conflicts is powerfully influenced by the magnitude of the movement of vertical time and the extent to which disputants are socially superior or inferior, dependent or independent, and intimate or unacquainted.
Turning to the controls, respondents were more apt to contest conflicts that made them especially angry (as compared to low anger). Respondents were also quicker to fight in response to creativity conflicts and validity conflicts than productivity conflicts: For all the emphasis on “publish or perish” in academic life, scholars appear to care more deeply about the origin and merit of ideas. Interestingly, the respondent’s self-described level of disputatiousness at work did not matter. Our measure of disputatiousness is imperfect as respondents might not be the best judge of their own tendencies. Nonetheless, the measure offers a legitimate indicator if respondents have a reasonable level of self-awareness. If so, then the impact of personality has been overestimated—contentious social locations appear to be more fateful than contentious people. 14 Additionally, prior work suggests that women should be slower to pursue grievances in the workplace (Gwartney-Gibbs and Lach 1994), but our data do not support that prediction: Contestation occurred in 40 percent of female-male conflicts compared to 41 percent of female-female conflicts, 40 percent of male-female conflicts, and 39 percent of male-male conflicts. The main effect of the respondent’s gender was similarly nonsignificant. The race of the respondent did not affect contestation either, although, as mentioned earlier, the number of conflicts involving faculty of color was small, perhaps explaining the null pattern. Nor did the holy grail of tenure influence contestation, suggesting that it is not simply the status of the aggrieved party that matters but instead the status of the aggrieved party relative to the status of the adversary. 15
Discussion
Long the subject of much casual observation and comment, intellectual conflict among academics has attracted little sustained scholarly attention. When it has, high-status cases have garnered disproportionate interest. For instance, an entire book was written about a 10-minute tussle between a pair of philosophers (Edmonds and Eidinow 2001). The adversaries were two of the most renowned thinkers of the twentieth century: Karl Popper and Ludwig Wittgenstein.
Much less is known about conflicts among competent but not elite academics. Our interviews at two contrasting universities shed some initial light on how the academic middle class handles intellectual conflicts.
We draw on a tried and tested theory of conflict management—Blackian theory—to explain variation in the contestation of intellectual conflicts. Combining synchronic and diachronic aspects of social geometry, we argue that assertiveness increases with the social distance and superiority of the aggrieved party and the movement of vertical time. Intellectual conflicts that span greater functional and relational distances, are directed downward in vertical space, and are larger movements of vertical time are handled more forcefully. Contestation is therefore most likely in conflicts over ideas that involve possible or actual loss of status and where the aggrieved party is of higher status, independent of, and a stranger to the other party. Conversely, socially close and upward conflicts involving small movements of vertical time are generally handled with considerable restraint. In that sense, intellectual conflict management behaves much like other forms of everyday conflict management (see e.g., Baumgartner 1998; Cooney 2009a; Hoffmann 2006).
Some expected patterns did not emerge in our data. Thus, the social characteristics of the aggrieved parties—their race, gender, and tenure status—had no effect. Perhaps more refined measures of the escalation of conflict would reveal otherwise. Similarly insignificant in our analysis was a popular commonsensical explanation: personality. Those who rated themselves as more aggressive in pursuing conflict did not consistently pursue conflict more vigorously. Although our self-report measure of contentiousness is an imperfect measure of personality, our data suggest that academics tend to over-psychologize social life because not a single respondent contested every grievance. While there may be aggressive people, there are also aggressive social geometries that appear to be even more fateful to the handling of conflict. When the social geometry for contestation exists, people tend to contest—even those who prefer to shy away from conflict. When the social geometry for restraint exists, people tend to exercise restraint—even those who are reluctant to let grievances slide.
Our findings are also inconsistent with the reputation that the academy sometimes has of being highly contentious. Certainly, many academics can recount war stories—often bitter—at the departmental, school, university, or disciplinary level. Knowledge of these conflicts sometimes jumps the walls of the university and, especially when reported and discussed in the media, serves to disseminate the perception of academic combativeness (see e.g., Eakin 2013; Ohlheiser 2013; Reddaway and Cohen 2012). The impression that academia is a site of incessant bickering results from selecting on the dependent variable—focusing on the highly public conflicts that do occur. We sought to minimize this problem by asking our respondents to recount not the most assertive action they had ever taken in response to an intellectual conflict but rather how they had responded to the most serious intellectual conflicts they had experienced in their careers. With this expanded view of nonpublic or private disputes, the most striking thing about intellectuals in the modern midrange American research university is not how much they fight but rather how little they fight. Most academics handle their intellectual conflicts with marked restraint. Their most common response to an intellectual grievance is to do nothing—to tolerate the offense. Any action they do take is generally mild, such as avoiding or lightly confronting the other party. This restraint occurs despite what is often significant provocation such as having ideas stolen, being elbowed down or off the list of authors, having work stonewalled by rivals, or receiving incompetent or hostile reviews. However, restraint is not typically a product of goodwill toward those toiling in the same vineyard. Intellectuals often exercise restraint toward each other in public while, as our research participants mentioned, grumbling or gossiping about each other in private. Restraint, rather, is a product of the social geometry of conflict—its location, direction, and change in social space.
The oft-cited maxim, “The politics of the university are so intense because the stakes are so low” appears, then, to be wrong, at least as regards intellectual conflict, for two reasons. 16 Not only is most intellectual conflict in the university typically not intense, the stakes are not invariably low. As in all social fields, some conflicts are consistently treated as unimportant (e.g., non-citation of peripheral references). But to intellectuals, there is nothing trivial about a scientist having a brilliant research idea stolen by another who uses it to win greater renown, a scholar being denied tenure because of her heterodox views, or a creative theorist having his work stonewalled by the discipline’s gatekeepers: These are all significant movements of vertical time.
Our treatment of the social geometry of intellectual conflict is far from exhaustive. We have analyzed the geometry of particular conflicts but not the larger social geometry of the modern American university. Certain features of that geometry encourage restraint. For instance, academia generally has low levels of institutional mobility. To change an academic post is a lengthy and uncertain process involving a host of professional, personal, and familial considerations. As Hirschman (1970:86) notes, open complaining or “voice” is more effective when the possibility of exit is greater. Those who cannot leave will confront less, in part because confronting may only make things worse in the future. Other geometric aspects of academic life encourage contestation such as the promulgation of new ideas: both the criticism of ideas as not being new enough and the tendency to reject ideas as being too new. Innovation and its rejection is a common source of conflict in all cultural fields (Black 2011). Future work could profitably take account these general geometric features.
Conclusion
Intellectual conflict is a topic ripe for further exploration. Our initial treatment has focused on a single, albeit crucial issue: contestation. Academics, it appears, contest only a minority of conflicts. Contestation is not random or arbitrary but geometric. Academics are more likely to contest intellectual conflicts that emanate from more distant regions of relational and functional space, from lower in the status hierarchy, and that represent a larger fall in social status, threatened or actual. Contestation is, however, only one possible issue of importance. In the approximately 150 hours of interviews we conducted, a host of issues was raised either by what the respondents said or did that could warrant investigation by other scholars. Consider briefly four such issues.
Given the negative personal consequences of being embroiled in controversy, a number of our respondents mentioned techniques they have devised for minimizing or avoiding conflict. These include spatial strategies—avoiding the adversary as much as possible—and temporal strategies—waiting before responding to a grievance, whether virtually or actually. One participant adopts what might be called a policy of mindful ignorance, deliberately not attempting to figure out who has written highly negative reviews of her work. Another never talks professionally about his work until it is published—that way, it cannot be stolen. A third cited with approval his former major professor’s dictum that “the easiest way to make enemies is not to cite someone’s work.”
Reputation and its importance was a topic that surfaced in several interviews (see Black 1976). Formed and consolidated through discussions that take place in the absence of the person concerned, reputations can have fateful consequences. Having a name for being fair, easy to work with, and doing what you promise were desirable qualities mentioned by some of our research scientists. Particularly damaging was a reputation for intellectual theft. A specialist in education refused an editorship because one of the associate editors was someone who was known for taking other people’s ideas, changing them slightly, and then passing them off as his own. A professor of business has learned to avoid the ostensibly friendly approaches of a senior person in the field because she believes that person previously appropriated a research idea presented at a research workshop. On the other hand, one of our laboratory scientists noted that if the culprit is of sufficient eminence in his field, a known tendency to appropriate the good ideas of others may be excused. In the particular case, a renowned researcher muscled his way onto a project and up the author list via the review process, a move the respondent regarded as unethical but which his co-authors were inclined to dismiss simply as the hyper-competitiveness of elite science, particularly after it resulted in a high-status publication.
Several respondents expressed frustration at the standards of intellectual achievement in their disciplines, but there was a significant humanist-scientist divide. Humanists were critical of the lack of clear criteria for evaluating work. A literature specialist, for example, bemoaned the subjectivity of standards that dominate his field, stating that textual interpretations “simply fly past each other” and are “religious in nature.” A historian of music felt that the most recognition does not go to who has the best evidence but who has the most powerful presence in person and in print. Scientists, on the other hand, were more apt to complain about the unusually high demands on their time a successful career makes, creating challenging work-life choices. And for some, the arbitrariness of the reward system was a cause for considerable concern. With so many scientists pursuing so relatively few research dollars, which grants get funded seems to turn to a significant degree on status or chance. As a neuroscientist put it, science is suffering from a crisis of over-competence.
Our research addresses the handling of intellectual conflicts as an end it itself rather than their consequences for intellectual life. Even so, it is clear that conflicts over ideas can and do affect whether and how ideas get produced and disseminated. In some cases, respondents let lapse papers into which they had put considerable effort because they came to distrust the integrity of collaborators. For instance, a life sciences researcher expressed regret that he had not been able to form a research collaboration with a colleague in another department after they had a sharp disagreement over the administration of a previous grant. In other cases, respondents stood by and watched higher status academics appropriate ideas that revitalized their careers. A respondent whose research idea was usurped by a more prestigious scientist abandoned that field of inquiry as he felt that the other person would be willing and able to block the publication of his work in visible outlets. 17
Beyond our interview data, many other important issues remain to be explored. For instance, the critical question of the conditions under which some intellectual disputes erupt into public feuds between warring camps cannot yet be answered. Similarly awaiting investigation is the handling of conflict among intellectual elites—academics at the most prestigious institutions. Finally, intellectual conflict beyond academia—among artists, inventors, professionals of various stripes, and in everyday life—warrants study in its own right. In short, the handing of intellectual conflict is long overdue for systematic empirical investigation; this paper simply demonstrates the potential of social geometry to shape a central feature of the life of the mind.
Footnotes
Acknowledgements
For providing comments on an earlier draft, we thank: Donald Black, Paul Colomy, Jared Del Rosso, Hava Gordon, Joseph Hermanowicz, Anne McKenna, and Barry Schwartz.
Authors’ Note
Authors are listed in alphabetical order to denote equal contribution.